Endogenous vitamin D levels and breast cancer risk
| First author (ref.), year, study place | Years of data collection | Study design | No. cases/controls | Comparison* | OR (95% CL) | Variables adjusted for |
|---|---|---|---|---|---|---|
| Hiatt (45), 1998, United States† | 1964/1972-1991 | Nested case-control study | 96/96 | Serum 1,25(OH)2D (pg/mL), ≥51 vs <32 | 1.0 (0.2, 3.4) | Education, parity, history of breast biopsy, alcohol, breast cancer in mother and sister |
| Janowsky (44), 1999, United States‡ | 1990-1991 | Hospital-based case-control study | 131/149 | Blood 1,25(OH)2D (pmol/mL), ≤33.61 vs >62.94 | 5.3 (2.1, 13.4) | Age, assay batch, month of blood draw, clinic, and sample storage time |
| Blood 25(OH)D | No association§ | |||||
| Lowe (49), 2005, United Kingdom | 1998-2003 | Hospital-based case-control study | 179/179 | Plasma 25(OH)D (nmol/L), <50 vs >150 | 5.83 (2.31, 14.7) | Matching variables, time of year, age at sampling, and menopausal status |
| Bertone-Johnson (46), 2005, United States∥ | 1989/1990-1996 | Nested case-control study | 701/724 | Plasma 25(OH)D (quintile), Q5 vs Q1 | 0.73 (0.49, 1.07) | Matching variables, BMI at age 18, parity/age as first birth, family history of breast cancer, history of BBD, HRT, age at menarche, age at menopause, alcohol intake, and plasma α-carotene |
| Plasma 1,25(OH)2D (quintile), Q5 vs Q1 | 0.76 (0.52, 1.11) |
Abbreviation: OR, odds ratio.
↵* The associations shown are for all subjects combined. No associations by menopausal status were reported by these studies.
↵† This study was nested in a cohort of 95,000 women. During an average 15.4-year follow-up, 2,131 new breast cancer patients were identified from the cohort. Ninety-six White cases and 96 White controls were randomly selected from study subjects with serum collected in 1964 to 1972.
↵‡ Blood samples were collected after diagnosis. Results presented here are for White women only because sample sizes for other ethnic groups were too small.
↵§ The article showed that the blood 25(OH)D level did not differ between cases and controls. However, the article did not present adjusted OR and 95% CL for the association between blood 25(OH)D level and breast cancer risk.
↵∥ This study was nested in the Nurses' Health Study.